Pith. sign in

REVIEW 1 cited by

When Explanations Lie: Why Many Modified BP Attributions Fail

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1912.09818 v7 pith:HCRIZEKH submitted 2019-12-20 cs.LG stat.ML

classification cs.LGstat.ML
keywords methodsmodifieddeepliftempiricallyanalyzeexplanationslaterlayers
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Attribution methods aim to explain a neural network's prediction by highlighting the most relevant image areas. A popular approach is to backpropagate (BP) a custom relevance score using modified rules, rather than the gradient. We analyze an extensive set of modified BP methods: Deep Taylor Decomposition, Layer-wise Relevance Propagation (LRP), Excitation BP, PatternAttribution, DeepLIFT, Deconv, RectGrad, and Guided BP. We find empirically that the explanations of all mentioned methods, except for DeepLIFT, are independent of the parameters of later layers. We provide theoretical insights for this surprising behavior and also analyze why DeepLIFT does not suffer from this limitation. Empirically, we measure how information of later layers is ignored by using our new metric, cosine similarity convergence (CSC). The paper provides a framework to assess the faithfulness of new and existing modified BP methods theoretically and empirically. For code see: https://github.com/berleon/when-explanations-lie

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Rethinking Neural Width for Alternating Current Optimal Power Flow Proxies

    cs.LG 2026-06 unverdicted novelty 5.0 of 10

    LG-ND algorithm finds neural networks with up to 10x fewer neurons per layer that match baseline performance on ACOPF approximation for IEEE systems.

Pith tools